Meep originated as part of graduate research at MIT in the mid 2000s with initial contributions by Steven G. Johnson, Ardavan Oskooi, David Roundy, Mihai Ibanescu, and Peter Bermel. The project has been under continuous development for nearly 20 years. Currently, the Meep project is maintained by an active developer community on GitHub. Christopher Hogan and M.T. Homer Reid lead the development of the Python interface, mode-decomposition feature, and GDSII import routines. M.T. Homer Reid and Alec Hammond developed the adjoint solver. Alex Cerjan assisted with adding support for saturable absorption via multilevel atomic gain media. Alec Hammond developed the visualization module. Yidong Chong and Alex Cerjan added support for gyrotropic media. Andreas Hoenselaar contributed to several performance enhancements. Krishna Gadepalli added support for checkpointing the simulation state.


We request that you cite the following technical reference in any work for which you used Meep:

If you use Meep's adjoint topology-optimization facilities, you should additionally cite:

If you want a one-sentence description of the algorithm for inclusion in a publication, we recommend something like:

  • "Simulations were performed with the finite-difference time-domain (FDTD) method [ref FDTD], using an open-source software package [ref Meep]."

General references on the FDTD method include, for example:

Financial Support

Meep's development has been supported by Small Business Innovation Research (SBIR) Phase 1 and 2 awards from the National Science Foundation (NSF) under award numbers 1647206 and 1758596. Initial development was supported in part by the Materials Research Science and Engineering Center program of the NSF under award numbers DMR-9400334 and DMR-0819762, by the Army Research Office through the Institute for Soldier Nanotechnologies under DAAD-19-02-D0002, and DARPA under N00014-05-1-0700 administered by the Office of Naval Research.